1. Field of the Invention
The present invention relates to an ink jet recording method and apparatus. More particularly, the present invention relates to an ink jet recording method and apparatus capable of reducing streaks and unevenness in recording density and forming an image with good quality.
2. Description of the Related Art
An ink jet (IJ) recording method is widely used because of advantageous features, i.e., a high speed, high image quality, and a low running cost. Recently, with the development of information infrastructures such as computer hardware/software and networks, a higher speed and higher image quality have been demanded.
In ink jet recording techniques, one problem to be overcome for achieving higher image quality is a variation in ink droplets discharged from an ink jet recording head. Such a variation occurs, for example, in the amount of ink discharged and the discharge direction of ink droplets. It is also known that the variation in ink discharge causes the following problems.
(1) For a plurality of discharge ports of the recording head, if there is a variation in the discharge direction of ink droplets for each of the discharge ports, the positions of dots formed on the recording paper are offset and, hence, streaks appear in a recorded image.
(2) If there is a variation in the amount of discharged ink for each of the discharge ports, the sizes and densities of dots formed on the recording paper are varied and, hence, density unevenness appears in a recorded image.
(3) If there are variations in the direction and the amount of ink discharged through one discharge port, the positions, sizes, and densities of dots formed in the main scanning direction on the recording paper are varied; thus, a line in the main scanning direction may not be uniform, or a recorded image may have a rough feel when viewed. This problem is more likely to occur when using a bubble jet method, in which ink is discharged by utilizing thermal energy, instead of the other various discharge methods used in ink jet recording techniques.
As one method for avoiding those problems, the recording head is manufactured at very high accuracy to minimize variations in the direction and the amount of discharged ink for each of the discharge ports. However, such a method gives rise to problems that the production cost is increased and the production yield is reduced.
Also, as a method for eliminating density unevenness through software, Japanese Patent Laid-Open No. 57-41965, Japanese Patent Publication No. 2708439, Japanese Patent Publication No. 2711011, etc. disclose a process of changing the number of driven ink droplets so as to cancel variations in the amount of discharged ink among the discharge ports. The inventions disclosed in those related publications are concerned with the technique of recording a test pattern by using a recording head beforehand, reading the density of the recorded test pattern, and correcting the density unevenness. For example, information regarding characteristics, such as the ink discharge amount of the recording head, is obtained based on variations in the read density unevenness, and the obtained information is utilized in image processing executed when an image is actually recorded. The number of driven ink droplets, the amount of discharged ink, etc. are thereby adjusted so as to suppress the occurrence of streaks and density unevenness. The disclosed method is to execute a correction process by utilizing information of the density unevenness that is obtained from the result of recording the test pattern beforehand, and it is confirmed as being an effective manner for reducing the density unevenness.
However, the disclosed method has a problem as follows. When variations in the amount of discharged ink among the discharge ports change with time, the correction process is executed whenever the change occurs, by repeating the steps of recording the test pattern and reading the recorded test pattern. Accordingly, a plurality of processing steps must be carried out whenever the correction process is executed, and maintainability of the device deteriorates. In other words, the correction process requires not only time and labor for recording the test pattern, but also ink and recording paper for executing the process. Furthermore, when a user has recorded an image without noticing a reduction in image quality caused by deterioration with time, the ink, the recording paper, and the time required for the recording are wasted. In addition, the disclosed method is not effective when the direction and the amount of ink discharged through one discharge port vary for each of the discharge ports.
To overcome the problems mentioned above, Japanese Patent Laid-Open Nos. 60-107975 and 3-231861, for example, disclose a recording method of forming one line in the main scanning direction by ink droplets discharged through a plurality of discharge ports to reduce variations in the direction and the amount of discharged ink among the discharge ports so that streaks and density unevenness are less recognizable.
The disclosed method will be described below with reference to FIG. 1. In a preceding main scan of a recording head (“main scan” is referred to simply as “scan” hereinafter), pixels not adjacent to each other in both horizontal and vertical directions (i.e., the main scan direction of the head and a sub-scan direction of paper feed) are recorded. Then, recording paper is fed in the sub-scan direction through a distance corresponding to a half of the length of a discharge port row. In a subsequent scan, the remaining pixels which have not been recorded in the preceding scan are recorded.
In the example of FIG. 1, pixel data (image data), denoted by 1a, forming all pixels of each line is recorded by a method as represented in 1b. More specifically, all dots of each line are divided into a group of dots recorded in a preceding scan and another group of dots recorded in a subsequent scan. Then, the dots of the same line are recorded by using different recording elements (nozzles) in the preceding scan and the subsequent scan. With this method, taking the pixel data (image data) denoted by 1a in FIG. 1 as an example, each dot row (line) 1c in the main scanning direction is recorded by two different sets of discharge ports, whereby variations in the discharge direction are averaged and therefore streaks become less recognizable. Further, with the method shown in FIG. 1, assuming that a variation in the amount of discharged ink among the discharge ports occurs in normal distribution of standard deviations, a variation in the amount of discharged ink among the lines is reduced to σ/√2. Thus, because a variation in the amount of discharged ink among the lines is recognized as a variation in density, an image with less density unevenness can be obtained.
However, the method shown in FIG. 1 has a problem as follows. When recording a particular half-tone image (in which pixels are not all formed), for example, as denoted by 2a in FIG. 2, each dot row in the main scanning direction is formed by ink droplets discharged through the same sets of discharge ports according to that method. Therefore, the effect of reducing streaks and density unevenness is not obtained in such a case.
Further, since the method shown in FIG. 1 is intended to reduce streaks and density unevenness based on statistics and probability approaches by employing ink droplets from plural sets of discharge ports to average the directions and the amounts of discharged ink, it is not ensured that streaks and density unevenness are always reduced. For example, if the discharge directions of ink droplets from two discharge ports forming the same line are both curved in the same direction, or if the amounts of ink discharged through the two discharge ports are both reduced, the effect of reducing streaks and density unevenness cannot be obtained. In addition, the method shown in FIG. 1 is not sufficiently effective when the direction and the amount of ink discharged through one discharge port vary for each of the discharge ports.
Another recording method for making streaks and density unevenness less recognizable is disclosed in U.S. Pat. No. 5,430,479 filed by the inventors of this application. This method is intended to reduce streaks and density unevenness in a multi-value image by employing a multiplicity of ink droplets discharged through a multiplicity of discharge ports to form one pixel.
While the method disclosed in the above-cited Japanese Patent Laid-Open Nos. 60-107975 and 3-231861 performs binary recording, the method disclosed in U.S. Pat. No. 5,430,479 performs multi-value recording. The latter method is superior in that, even when recording a particular half-tone image, streaks and density unevenness are less apt to occur.
Even with the method disclosed in the above-cited U.S. Pat. No. 5,430,479, however, since streaks and density unevenness are reduced based on statistics and probability approaches by employing ink droplets from a plurality of discharge ports to average the directions and the amounts of discharged ink, it is also not ensured that streaks and density unevenness are always reduced.
In addition, the method disclosed in the above-cited U.S. Pat. No. 5,430,479 is also not sufficiently effective when the direction and the amount of ink discharged through one discharge port vary for each of the discharge ports.
A method for solving the above-mentioned problems is disclosed in U.S. Pat. No. 6,547,361. This method comprises the steps of reading a recorded image, comparing the read recorded image with the image information used to record it, determining positions of recording failures, and recoding again the same image while correcting the recording failures in a subsequent scan. As a result, streaks and density unevenness can reliably be reduced for any image. Further, even when the direction and the amount of ink discharged through one discharge port vary for each of the discharge ports, it is possible to reliably reduce streaks and density unevenness.
Regarding U.S. Pat. No. 6,547,361, this patent discloses the method for locating the positions of recording failures and performing correction recording in a subsequent scan, but does not sufficiently disclose the process of creating correction data that is effective to satisfactorily reduce streaks and density unevenness in addition correcting the recording failures.